To identify the influencing factors of delayed T-tube removal (>2 months) in patients with recurrent intrahepatic bile duct stones, and to establish a Nomogram prediction model.

Clinical data of 180 patients with recurrent intrahepatic bile duct stones who underwent surgery and retained T-tube drainage in the First Affiliated Hospital of Anhui Medical University from January 2018 to December 2021 were retrospectively analyzed. The informed consents of all patients were obtained and the local ethical committee approval was received. Among them, 54 patients were male and 126 female, aged from 15 to 85 years, with a median age of 57 years. Univariate and multivariate Logistic regression analyses were performed for the influencing factors of delayed postoperative T-tube removal. Based on the obtained independent risk factors, Nomogram prediction model was established. The prediction model was evaluated by the consistency index (C-index), calibration curve verification and ROC curve.

The median time of T-tube indwelling was 2(1,3) months. T-tube removal was delayed in 70 cases after operation. Multivariate Logistic regression analysis showed that delayed removal of intrahepatic stones, bilateral distribution of intrahepatic stones, intrahepatic bile duct steonosis, preoperative elevated WBC, TB≥34.2 μmol/Lon postoperative 1 day and positive results of bile culture were the independent risk factors for delayed T-tube removal in patients with recurrent intrahepatic bile duct stones (HR=2.609, 2.684, 5.250, 4.794, 4.035, 2.991; P<0.05). The C-index of the Nomogram prediction model constructed based upon these factors was 0.800 (95%CI: 0.735-0.866), and the optimal cut-off value of total score was 134. The sensitivity and specificity of the Nomogram model were 0.56 and 0.88, respectively.

The risk factors of delayed T-tube removal in patients with recurrent intrahepatic bile duct stones include delayed removal of intrahepatic stones, bilateral distribution of intrahepatic stones, intrahepatic bile duct stenosis, bile duct inflammation and injury. The Nomogram prediction model can preliminarily predict the short-term postoperative recovery of patients.